An investigation of fluvial sediments of the Middle Loup River at Dunning, Nebr., was begun in 1946 and expanded in 1949 to provide information on sediment transportation. Construction of an artificial turbulence flume at which the total sediment discharge of the Middle Loup River at Dunning, Nebr., could be measured with suspended-sediment sampling equipment was completed in 1949. Since that time. measurements have been made at the turbulence flume and at several selected sections in a reach upstream and downstream from the flume. The Middle Loup River upstream from Dunning traverses the sandhills region of north-central Nebraska and has a drainage area of approximately 1,760 square miles. The sandhills are underlain by the Ogallala formation of Tertiary age and are mantled by loess and dune sand. The topography is characterized by northwest-trending sand dunes, which are stabilized by grass cover. The valley floor upstream from Dunning is generally about half a mile wide, is about 80 feet lower than the uplands, and is composed of sand that was mostly stream deposited. The channel is defined by low banks. Bank erosion is prevalent and is the source of most of the sediment load. The flow originates mostly from ground-water accretion and varies between about 200 and 600 cfs (cubic feet per second). Measured suspended-sediment loads vary from about 200 to 2,000 tons per day, of which about 20 percent is finer than 0.062 millimeter and 100 percent is finer than 0.50 millimeter. Total sediment discharges vary from about 500 to 3,500 tons per day, of which about 10 percent is finer than 0.062 millimeter, about 90 percent is finer than 0.50 millimeter, and about 98 percent is finer than 2.0 millimeters. The measured suspended-sediment discharge in the reach near Dunning averages about one-half of the total sediment discharge as measured at the turbulence flume. This report contains information collected during the period October 1, 1948, to September 30, 1952. The information includes sediment discharges; particle-size analyses of total load, of measured suspended sediment, and of bed material; water discharges and other hydraulic data for the turbulence flume and the selected sections. Sediment discharges have been computed with several different formulas, and insofar as possible, each computed load has been compared with data from the turbulence flume. Sediment discharges computed with the Einstein procedure did not agree well, in general, with comparable measured loads. However, a satisfactory representative cross section for the reach could not be determined with the cross sections that were selected for this investigation. If the computed cross section was narrower and deeper than a representative cross section for the reach, computed loads were high; and if the computed cross section was wider and shallower than a representative cross section for the reach, computed loads were low. Total sediment discharges computed with the modified Einstein procedure compared very well with the loads of individual size ranges and the measured total loads at the turbulence flume. Sediment discharges computed with the Straub equation averaged about twice the measured total sediment discharge at the turbulence flume. Bed-load discharges computed with the Kalinske equation were of about the right magnitude; however, high computed loads were associated with low total loads, low unmeasured loads, and low concentrations of measured suspended sediment coarser than 0.125 millimeter. Bed-load discharges computed with the Schoklitsch equation seemed somewhat high; about one-third of the computed loads were slightly higher than comparable unmeasured loads. Although, in general, high computed discharges with the Schoklitsch equation were associated with high measured total loads, high unmeasured loads, and high concentrations of measured suspended sediment coarser than 0.125 millimeter, the trend was not consistent. Bed-load discharges computed